Table 4 Lessons learned from discontinuation of the PACE Plus trial and corresponding recommendations for future research

 • Even though treatment distribution follows legislation and works on paper, they may have issues in practice that affect patients.

 • Asking GPs to recruit incident cases during the first consultation seems unlikely to be successful considering the current workload in general practice.

 • Attempting to answer several research questions at once not only requires more patients but is also more complicated to explain to GPs and potential participants.

 • Interests and expectations of patients can collide with scientifically interesting questions in practice.

 • The number of available patients meeting inclusion criteria is easily overestimated (Lasagna’s Law).

 • Negative perception of trial treatment may influence participation of both GPs and patients.

• Keep treatment distribution as simple as possible and provide an attractive alternative to conventional therapy.

• Try to answer your research question in prevalent cases or use alternative designs such as a trial within a cohort study. Take into account reimbursement of GPs in grant application and budgeting (especially if you do end up recruiting incident cases).

• Choose your most important research question and design your trial to be as simple as possible.

• Before starting a trial, ask a patient panel for their preferences and expectations of treatments and outcomes. Ask GPs if they know of any reservations about treatments you consider using.

• Take Lasagna’s law into account when planning your trial.

• Consider conducting a pilot trial and taking part in Mandatory Continuous Education.